In this paper we give a simple account of local computation of marginal probabilities when the joint probability distribution is given in factored form and the sets of variables involved in the factors form a hypertree. Previous expositions of such local computation have emphasized conditional probability. We believe this emphasis is misplaced. What is essential to local computation is a factorization. It is not essential that this factorization be interpreted in terms of conditional probabilities. The account given here avoids the divisions required by conditional probabilities and generalizes readily to alternative measures of subjective probability, such as Dempster-Shafer or Spohnian belief functions.
This is a preview of subscription content, access via your institution.
Buy single article
Instant access to the full article PDF.
Price excludes VAT (USA)
Tax calculation will be finalised during checkout.
S. Arnborg, D.G. Corneil and A. Proskurowski, Complexity of finding embeddings in a k-tree, SIAM J. Algebraic and Discrete Meth. 8 (1987) 277–284.
C. Beeri, R. Fagin, D. Maier and M. Yannakakis, On the desirability of acyclic database schemes, J. ACM 30 (1983 479–513.
C. Berge,Graphs and Hypergraphs, translated from French by E. Minieka (North-Holland, 1973).
U. Bertele and F. Brioschi,Nonserial Dynamic Programming (Academic Press, 1972).
L.S. Brownston, R.G. Farrell, E. Kant and N. Martin,Programming Expert Systems in OPS5: An Introduction to Rule-Based Programming (Addison-Wesley, 1985).
C. Cannings, E.A. Thompson and M.H. Skolnick, Probability functions on complex pedigrees, Advan. Appl. Probab. 10 (1978) 26–61.
R. Dechter and J. Pearl, Tree-clustering schemes for constraint processing, in:Proc. 7th National Conf. on AI (AAAI-88), St. Paul, MN (1988) vol. 1, pp. 150–154.
A.P. Dempster, Construction and local computation aspects of network belief functions, Research Report S-125, Department of Statistics, Harvard University, Cambridge, MA (1988).
M.C. Golumbic,Algorithmic Graph Theory and Perfect Graphs (Academic Press, 1980).
D. Hunter, Parallel belief revision, in:Proc. 4th Workshop on Uncertainty in AI, Minneapolis, MN (1988) pp. 170–176.
C.W. Kelly III and S. Barclay, A general Bayesian model for hierarchical inference, Organizational Behavior and Human Performance 10 (1973) 388–304.
A. Kong, Multivariate belief functions and graphical models, Ph.D. Dissertation, Department of Statistics, Harvard University, Cambridge, MA (1986).
S.L. Lauritzen, T.P. Speed and K. Vijayan, Decomposable graphs and hypergraphs, J. Austral. Math. Soc. A36 (1984) 12–29.
S.L. Lauritzen and D.J. Spiegelhalter, Local computations with probabilities on graphical structures and their application to expert systems (with discussion), J. Roy. Statist. Soc. B50 (1988) 157–224.
D. Maier,The Theory of Relational Databases (Computer Science Press, 1983).
F.M. Malvestuto, Decomposing complex contingency tables to reduce storage requirements, in:Proc. 1986 Conf. on Computational Statistics (1986) pp. 66–71.
R.J. Meinhold and N.D. Singpurwalla, Understanding the Kaiman filter, Am. Statist. 37 (1983) 241–288.
K. Mellouli, On the propagation of beliefs in networks using the Dempster-Shafer theory of evidence, Ph.D. Dissertation, School of Business, University of Kansas, Lawrence, KS (1987).
J. Pearl, Fusion, propagation and structuring in belief networks, Artificial Intelligence 29 (1986) 241–288.
D.J. Rose, Triangulated graphs and the elimination process. J. Math. Anal. Appl. 32 (1970) 597–609.
D.J. Rose, A graph-theoretic study of the numerical solution of sparse positive definite systems of linear equations, in:Graph Theory and Computing, ed. R.C. Read (Academic Press, 1973) pp. 183–217.
R. Seidel, A new method for solving constraint satisfaction problems, in:Proc. 7th Int. Joint Conf. on AI (IJCAI-81), Vancouver, British Columbia, Canada (1981) vol. 1, pp. 338–342.
R.D. Shachter and D. Heckerman, A backwards view for assessment, AI Magazine 8 (1987) 55–61.
G.R. Shafer and P.P. Shenoy, Local computation in hypertrees, Working Paper no. 201, School of Business, University of Kansas, Lawrence, KS (1988).
G.R. Shafer, P.P. Shenoy and K. Mellouli, Propagating belief functions in qualitative Markov trees, Int. J. Approximate Reasoning 1 (1987) 349–400.
P.P. Shenoy, A valuation-based language for expert systems, Int. J. of Approximate Reasoning (1989) 383–411.
P.P. Shenoy and G.R. Shafer, Propagating belief functions using local computations, IEEE Expert 1 (1986) 43–52.
P.P. Shenoy and G.R. Shafer, An axiomatic framework for Bayesian and belief-function propagation, in:Proc. 4th Workshop on Uncertainty in Artificial Intelligence, St. Paul, MN (1988) pp. 307–314.
P.P. Shenoy and G.R. Shafer, Constraint propagation, Working Paper no. 208, School of Business, University of Kansas, Lawrence, KS (1988).
P.P. Shenoy and G.R. Shafer, Axioms for discrete optimization using local computation, Working Paper no. 207, School of Business, University of Kansas, Lawrence, KS (1988).
P.P. Shenoy, G.R. Shafer and K. Mellouli, Propagation of belief functions: A distributed approach, in:Proc. 2nd Workshop on Uncertainty in AI, Philadelphia, PA (1986) pp. 249–260. Also in:Uncertainty in Artificial Intelligence, vol. 2, eds. J.F. Lemmer and L.N. Kanal (North-Holland, 1988) 325–335.
W. Spohn, Ordinal conditional functions: A dynamic theory of epistemic states, in:Causation in Decision, Belief Change, and Statistics, eds. W.L. Harper and B. Skyrms (D. Reidel Publishing Company, 1988) vol. 2, pp. 105–134.
R.E. Tarjan and M. Yannakakis, Simple linear time algorithms to test chordality of graphs, test acyclicity of hypergraphs, and selectively reduce acyclic hypergraphs, SIAM J. Computing 13 (1984) 566–579.
L. Zhang, Studies on finding hypertree covers for hypergraphs, Working Paper no. 198, School of Business, University of Kansas, Lawrence, KS (1988).
Rights and permissions
About this article
Cite this article
Shafer, G.R., Shenoy, P.P. Probability propagation. Ann Math Artif Intell 2, 327–351 (1990). https://doi.org/10.1007/BF01531015
- Probability propagation
- local computation
- construction sequence
- hypertree cover
- Markov tree
- parallel processing